Skip to content

Latest commit

 

History

History
1087 lines (584 loc) · 36.2 KB

File metadata and controls

1087 lines (584 loc) · 36.2 KB

Public Models

All models were compiled using Hailo Dataflow Compiler v2.18.0.


Object Detection


Link Legend


Key / Icon Description
Networks used by Hailo-apps.
S Source – Link to the model's open-source repository.
PT Pretrained – Download the pretrained model file (ZIP format).
HEF, NV12, RGBX Compiled Models – Links to models in various formats: - HEF: RGB format - NV12: NV12 format - RGBX: RGBX format
PR Profiler Report – Download the model's performance profiling report.

Coco


Network Name float mAP Hardware mAP FPS (Batch Size=1) FPS (Batch Size=8) Links Input Resolution (HxWxC) Params (M) OPS (G)
centernet_resnet_v1_18_postprocess 26.4 25.0 99.4 177 512x512x3 14.22 31.21
centernet_resnet_v1_50_postprocess 31.8 29.2 57.6 100 512x512x3 30.07 56.92
damoyolo_tinynasL20_T 42.8 42.3 124 268 640x640x3 11.35 18.02
damoyolo_tinynasL25_S 46.5 45.3 78.2 148 640x640x3 16.25 37.64
damoyolo_tinynasL35_M 49.7 47.6 51.4 89.2 640x640x3 33.98 61.64
detr_resnet_v1_18_bn 33.9 31.5 23.4 50.0 800x800x3 32.42 61.87
efficientdet_lite0 27.3 26.5 92.0 239 320x320x3 3.56 1.94
efficientdet_lite1 32.3 31.8 75.9 184 384x384x3 4.73 4
efficientdet_lite2 35.9 34.7 39.8 88.4 448x448x3 5.93 6.84
nanodet_repvgg 29.3 28.5 620 620 416x416x3 6.74 11.28
nanodet_repvgg_a12 33.7 32.5 201 201 640x640x3 5.13 28.23
nanodet_repvgg_a1_640 33.3 32.9 201 201 640x640x3 10.79 42.8
ssd_mobilenet_v1⭐ 23.2 22.4 356 356 300x300x3 6.79 2.5
ssd_mobilenet_v2 24.2 23.0 131 293 300x300x3 4.46 1.52
tiny_yolov3 14.6 14.3 899 899 416x416x3 8.85 5.58
tiny_yolov4 19.2 17.7 895 895 416x416x3 6.05 6.92
yolo26m 51.7 50.0 28.6 47.1 640x640x3 20.4 68.4
yolo26n 39.2 37.4 111 255 640x640x3 2.4 5.5
yolo26s 46.8 44.8 66.6 147 640x640x3 9.5 20.9
yolov10b⭐ 52.0 50.8 25.9 45.7 640x640x3 20.15 92.09
yolov10n⭐ 38.5 36.6 150 359 640x640x3 2.3 6.8
yolov10s⭐ 45.9 45.0 87.6 187 640x640x3 7.2 21.7
yolov10x⭐ 53.7 51.8 12.0 18.1 640x640x3 31.72 160.56
yolov11l⭐ 52.8 52.0 21.8 37.1 640x640x3 25.3 87.17
yolov11m⭐ 51.1 49.8 35.3 58.1 640x640x3 20.1 68.1
yolov11n⭐ 39.0 37.8 157 371 640x640x3 2.6 6.55
yolov11s⭐ 46.3 45.5 91.9 192 640x640x3 9.4 21.6
yolov11x⭐ 54.1 53.1 12.1 18.1 640x640x3 56.9 195.29
yolov3 38.4 38.3 15.7 19.6 608x608x3 68.79 158.10
yolov3_416 37.7 37.6 34.0 69.6 416x416x3 61.92 65.94
yolov3_gluon 37.3 35.7 25.6 39.9 608x608x3 68.79 140.7
yolov3_gluon_416 36.3 34.2 31.7 57.3 416x416x3 61.92 65.94
yolov4_leaky 48.3 47.1 22.1 32.0 512x512x3 64.33 91.04
yolov5m⭐ 42.6 41.3 60.4 105 640x640x3 21.78 52.17
yolov5m6_6.1 50.7 49.3 16.2 19.6 1280x1280x3 35.70 200.04
yolov5m_6.1 44.7 43.5 69.2 118 640x640x3 21.17 48.96
yolov5m_wo_spp⭐ 43.1 41.5 62.9 106 640x640x3 22.67 52.88
yolov5s⭐ 35.3 34.1 124 243 640x640x3 7.46 17.44
yolov5s_bbox_decoding_only 35.3 34.1 124 243 640x640x3 7.46 17.44
yolov5s_c3tr 37.1 35.7 120 240 640x640x3 10.29 17.02
yolov5s_wo_spp 34.8 33.8 137 275 640x640x3 7.85 17.74
yolov5xs_wo_spp 33.2 32.1 206 438 512x512x3 7.85 11.36
yolov5xs_wo_spp_nms_core 32.7 31.1 57.2   512x512x3 7.85 11.36
yolov6n⭐ 34.3 32.5 356 356 640x640x3 4.32 11.12
yolov6n_0.2.1 35.2 34.1 560 560 640x640x3 4.33 11.06
yolov6n_0.2.1_nms_core 35.2 34.0 199 199 640x640x3 4.32 11.12
yolov7⭐ 50.6 48.8 36.7 58.1 640x640x3 36.91 104.51
yolov7_tiny 37.1 36.1 157 297 640x640x3 6.22 13.74
yolov7e6 55.4 53.7 6.43 7.97 1280x1280x3 97.20 515.12
yolov8l⭐ 52.4 51.8 26.1 40.7 640x640x3 43.7 165.3
yolov8m⭐ 49.9 49.2 50.8 87.0 640x640x3 25.9 78.93
yolov8n⭐ 37.0 36.4 202 438 640x640x3 3.2 8.74
yolov8s⭐ 44.6 44.0 110 208 640x640x3 11.2 28.6
yolov8s_bbox_decoding_only 44.8 44.2 110 208 640x640x3 11.2 28.6
yolov8x⭐ 53.5 52.9 16.2 23.2 640x640x3 68.2 258
yolov9c⭐ 52.6 51.5 27.2 42.2 640x640x3 25.3 102.1
yolox_l_leaky 48.7 46.6 25.4 39.2 640x640x3 54.17 155.3
yolox_s_leaky 38.1 37.3 109 199 640x640x3 8.96 26.74
yolox_s_wide_leaky 42.4 41.0 64.0 102 640x640x3 20.12 59.46
yolox_tiny 32.6 31.3 214 510 416x416x3 5.05 6.44